Bringing On The Beeb
This article originally appeared in Outsource magazine Issue #30 Winter 2012
He’s where the buck stops for almost all the technology owned by the world’s biggest broadcasting organisation. It’s no surprise that BBC CTO John Linwood has his hands full. However, as he tells Outsource in this exclusive Q&A, the challenges are as fascinating as they are complex – and the opportunity to drive real value in even the most problematic of environments has never been greater…
Outsource: John, what are your biggest challenges right now?
John Linwood: The big challenges are that technology is changing faster and faster, regulations are changing, competitive landscapes are changing, the way people work is changing – and all of that has an impact on the technology that we deliver to the organisation. So how do we keep pace with that technological change? How do we keep the BBC on air and functioning and operating?
Then very specifically, how do we drive value for money? Because the licence fee is flat for the next four years now, so we have to deliver what we’ve always delivered for less in real money terms. So – as always with people in my role – cost saving is important, managing costs is important. Service delivery is absolutely critical and continues to be.
A third challenge is the BBC has a very, very large technology asset base. Technology is part of everything we do at the BBC, and so refreshing that asset base is a huge job just in itself.
What we’re trying to do at the same time is make the BBC more agile, leaner, more effective – and to maintain the quality of our output and our audience appreciation. So what that means specifically is that managing suppliers is absolutely critical to us in terms of keeping the technology going, dealing with in some cases very old legacy technology – and also trying to find new ways to deliver services without having to buy quite so much infrastructure or such large services going forward.
O: What’s your outlook on how best to achieve cost savings whilst maintaining that quality? It’s a challenge shared by many CTOs and CIOs, of course…
JL: Well, I think there are a number of ways. First of all, overall the BBC had to find 20 per cent of savings. Now you can’t find 20 per cent of savings by trimming: you can’t just go around taking little things out. But equally one of the things our audience tells us loud and clear is that they don’t want us to cut any services – and the best example of that was when we decided to shut down 6 Music, not only did its listenership triple overnight, but there was a huge public outcry. So, if we can’t cut services and we can’t go around trimming, then we have to find completely different ways of doing things.
What we have done is start to look more widely at the way in which we deliver the technology and services in the BBC. I’ll give you an example: one area we are doing a lot of work in is around centralisation – so moving a lot of our technology away from dedicated infrastructure and service onto virtualised infrastructure where you get greater value for money in terms of the infrastructure that you have, and greater output.
We’re looking at different ways of supporting things, how we manage technology remotely so we don’t have to have so many people out and about in the field. Then looking at our technology vendors – in fact all of our vendors – and saying; first of all how can they help us with their pricing, secondly how can they help us in terms of innovation and technology change that helps deliver new ways of doing things?
If you look at a typical studio today, it’s very different from a studio from 10 or 15 years ago, with lots more commodity standardised hardware, whereas 10-15 years ago you would have had more custom-built hardware and systems. Much more of it today is straight-off-the-shelf stuff.
O: You mentioned innovation there, and there are of course various different interpretations of what innovation even is: what’s your definition of innovation when it comes to dealing with your suppliers? And how are you ensuring that you get it?
JL: What we are looking for is suppliers who can help us reach outcomes rather than sell us technologies. I can give you a real example of that: as a broadcaster, as a news organisation, and as a content producer, networking is incredibly important to us, for content acquisition as well as content production, post-production and then eventual distribution. Historically, we use dedicated circuits for everything – so if you were at Glastonbury Festival we would have had a set of dedicated ISDN lines that would do at Glastonbury, then as we moved content around the BBC or put live feeds in, we would have huge numbers of dedicated circuits.
One way in which a supplier could provide innovation to the BBC is to say, well, all that could be moved into an IP network: we don’t need to have dedicated infrastructure, we could have a pay-per-use model. Think about something like sport. Sport has huge bandwidth requirements: on a Saturday afternoon, or early Saturday evening, there will be a lot of sport activity going on – but those circuits would be idle for most of the week. Under the current networking infrastructure you’d still be paying for them. Under a managed service, you would be only paying for them as you use them.
Those are the kinds of things that we have been working on with vendors. We’ve got a big programme going on at the moment to look at renewing our huge technology outsourcing contract and the first phase of that has been to work with a whole raft of vendors just to understand how they would deliver the technology today versus how they would have delivered eight years ago when we did the contract. The real challenge we are laying down to them is to say, the status quo is not the given; actually what we want to do is completely revamp it and build outcomes-based services where they can help us adapt, change, meet technological change and take advantage of any new innovation that would bring costs down. So when we talk about innovation, that’s some of what we’re talking about.
Obviously then there’s innovation in pure technology terms. If you look at the example of news gathering, today we use multi-bundled 3G devices connected to a camera to get live feeds where it’s not practical or possible to get a satellite link. Those are the kinds of innovations that help us get our job done. Then of course there’s massive innovation going on in the content production world today around digital production.
O: Talking about working with vendors: how many suppliers do you have? It must be a pretty significant number.
JL: Actually, it’s not a huge number because we have a primary contract with Atos that covers a good chunk of our technology delivery: probably in the region of half of our technology to the BBC comes through that one contract. Now, under that contract are many sub-contractors so, yes, if you want to expand it out, there are probably several dozen suppliers involved, but we only have one contract in place there.
We obviously have lots and lots of small suppliers who provide specialist or small services to us. But in terms of large suppliers, you could probably name all of them. We buy consulting from a number of different places, we buy technology from a number of different places, but the bulk of it comes through the Atos contract, the Steria contract, Capita as well. It’s quite a small number of contracts all-told.
O: Because of course another area where there’s been a lot of innovation in this space is around contract, and organisations moving to more complex multisourced environments: is that something you’re looking at as well?
JL: Yes. As I say, we have this one big contract: it was a ten-year contract, it was 1,000 pages long, it was very inflexible in its definition. Now we are doing a number of things. The first is, we’re moving to a multisource model. The reason is we want to have best-of-breed suppliers in different areas. Now, we’re not going to have hundreds of suppliers, we’re probably going to have five or six towers, something of that order. So we’ll take this one contract and break it up into five or six bundles.
The second thing we are doing is looking for contract flexibility and that means two things; one is flexibility in terms of shorter timeframes – so we’re unlikely to do ten-year contracts for most of them (sometimes we have to do longer contracts, so if you look at something like our HD distribution contract, it has to be much longer because the sheer cost of the infrastructure means unless you want to pay billions a year, you have to spread it over a long period of time to recoup the capital cost). But we are looking at variable length contracts, different contracts of different lengths. Some things where there’s high volatility we’ll go for shorter contracts and where there’s lower volatility we’ll go for longer.
The third thing we are looking at is what I mentioned earlier: outcomes-based contracts. For example, staff need to communicate with each other, but we’re not necessarily specifying it has to be mobile phones and email and unified comms, but saying these are the kinds of activities that staff need to be able to carry out. One of the classic examples is our existing contract says ten per cent of BBC staff should have access to WiFi. Now of course today that sounds ridiculous and in fact that isn’t the case – we have 100 per cent coverage of WiFi – but at the time, that was quite avant garde. Now what I don’t want to do is in five years’ time find that we’ve got another contract that’s completely out of date.
Another thing we’re doing is we’re going to stagger the procurement. It all comes to a head in March 2015, but we’re looking at perhaps doing it over a two-year period. We couldn’t, even if we wanted to, re-procure everything in one go and then transition it all in six months; it just isn’t feasible. So we’ll stagger the procurement over two years; that helps us, it helps the suppliers, it means we can focus on the bundle in hand – and it also means when we come to re-procurement in five, six, seven years’ time, again we won’t be on this cliff-edge of everything happening at the same time.
The final piece with the re-procurement, is that obviously we will have to have a service integration layer because we are multi-sourcing, so there’s a lot of work going on at the moment around defining that service integration layer – what does it look like? – then debating about who delivers that. So do we go out to the market and buy that as a service? If we do buy it as a service, can whoever delivers that also provide one or more of the towers? Or should we bring it in-house and have it managed in-house. Depending on who you talk to, you get different opinions on that, so we are doing some work to evaluate the various options.
O: Look at the CTO role more broadly: do you feel that, for example, supply management skills are becoming more important?
JL: Yes absolutely. Historically CIOs and CTOs had all of the technology under them and they hired everyone and they ran it all themselves and vendors provided software and hardware and some level of service… But clearly for the last ten years, more and more of that is outsourced. The second thing that has happened is, no longer can you buy all services from one company – so that you always end up in a mixed environment with multiple suppliers and integration challenges, and clearly suppliers have got better and better at being smart about contracts and service delivery, which is great on the one hand but it also means you’ve got to be sharper in managing them.
One of the key things is that you have to be a smart customer. You have to understand what you are buying; you have to understand what’s being delivered; and that’s becoming harder and harder because the complexity of what you’re buying and the complexity of what’s being delivered is becoming greater and greater. You mentioned supplier management; we have invested heavily in supplier management and that’s paid huge dividends because we’ve managed to drive greater value out of our suppliers and greater service levels out of our suppliers, but it means as a percentage of your overall spend, the amount you’re spending managing your suppliers is becoming a bigger and bigger slice.
Actually the BBC is probably below average in the industry. The industry will see between two and four per cent of the cost of the contract spent on supplier management. We’re not at that level yet but I suspect as we go through the next round of outsourcing, because we are going to a multisource environment and because the world has become more complex, a greater percentage of the money will be spent on supplier management. That’s a good thing! Because what it does is make sure you are getting value for money; it makes sure the suppliers are delivering what you want them to do and it also means that you can be a smart customer and have that knowledge in-house to ask the right questions, to buy the right things and to make sure they are delivering what you’ve paid for.
O: Tangentially: where do you sit on the bring-your-own-device issue?
JL: We are a huge supporter of it and we already have it. In all BBC buildings, there is a WiFi network that you as a visitor can connect to; there’s a WiFi network that staff can connect their own personal devices to; and then there’s a secure corporate network that only BBC devices can connect to. Really, the thinking is severalfold. First of all staff can bring in their own device and connect it to the network; secondly they can get to a number of the services (not all of them yet) and various applications on their own device. Then the third thing is, we’re doing a lot of work to not only let you use any device, but also any device from anywhere. So you don’t necessarily have to be in a BBC building to connect. I absolutely believe that the right way forward is to allow people to use the right device – and if they’re comfortable using their own personal device, then we should allow them to do that.
O: How are you looking to overcome what are some pretty hefty security challenges?
JL: A number of different ways. Clearly the issues around bring-your-own device are many – and actually the technology issues are the smallest of all of those. There are really two big areas: one is around security and one is around policies. On the security side, for example on Android devices we’re using Good Technology to provide a sandbox secure environment on the Android device. That does a few things for us – and it’s not unique to Good Technology, but it’s an example; it allows us first of all to have secure communications and secure storage on the device. So if the device gets lost nobody should be able to get into that.
Secondly it allows us to wipe that area of the device so we don’t have to wipe the whole device if somebody loses it. So if you brought your own personal device in and you had BBC stuff on there and then you lost it, you might say “don’t wipe my personal photos on there because it’s got a picture of my daughter and I don’t have another copy of that”; so using that technology we can wipe just the work piece of the device. The third thing it does is it allows us to secure a device with a second password, so that it means people can play with their phones, and let their kids play with them, without worrying about BBC content. If a staff member has BBC confidential content in their phone, the technology separates this into a secure area. That way if the person lets family or friends use the phone the BBC content is still secure.
Ironically, though, we fuss about these things, and we encrypt all of our laptops and all of our information on mobile devices – yet for years people walked around with bundles of paper which were totally unencrypted. So this isn’t a new problem; it’s just that there’s probably more data on a laptop or on a mobile device than was ever in a bundle of papers.
O: Talking about a whole load of data: can you give my readers a bit of background in terms of the challenge that maintaining the BBC archive entails and why it’s such an important issue?
JL: The BBC has been in business for 90 years, and has had great foresight in keeping huge amounts of material over that 90 years in every format you can think of, from shellac disks through to modern digital storage. There’s a number of things this does for us. First of all it’s a national treasure: it’s an archive of the history of the nation – and in fact the history of the Commonwealth nations and to some extent the world in many cases, with news and all sorts of events going back over time. We have a responsibility under the BBC’s Royal Charter to protect that for the nation. We’re just custodians of it for future generations. We have first of all to make sure that it’s kept safe, it’s not stolen or burnt or damaged. The second thing is it rots away, so we have to do preservation work on it – and we have many different formats. We have six million photographs, we have huge amounts of paper, through to film archives and then all of the radio and television archives over time. Preservation is actually quite a complex one; with the film archive we had to wash all of the nitrates out of the film to make it safe, to stop it getting vinegar syndrome. On a lot of the audio tape, the substrate that the tape is on is cellulose and it basically rots, like leaves. So preservation is a huge piece of work for us.
But then the second part of it is to say, how do we open that archive up and make it accessible? Today it’s accessible to BBC staff who use it for inclusion in programmes. We send out something like 5,000 tapes a week from the archive to be used mainly on BBC productions, but other productions outside the BBC as well. Also it’s available to historians and researchers, so we work with the British Film Institute and the British Library who provide a sort of front-end to our archive for people doing research work.
But that doesn’t really address the general public; it doesn’t allow you as an individual to come along and say “I want to see that episode of Blue Peter that my mother was on or I was on as a kid”. The next stage of that is to look at ways in which we can bring more and more of that content to the general public. Clearly where there is direct commercial value to the content, we digitise that because it has commercial re-sale value and we use it. We’re also looking, as we’re going through our preservation, to digitise more and more of that content and find ways in which we can deliver that to our audience. That may range from ways in which people can find out what was on television in 1962, through to seeing some of the actual content and having it available on the web.